The relationship between business value and technology is not a direct one. This may seem obvious, but for many organisations, the trap of ‘improve the technology, improve the business’ is too easy to fall into.
Technology is at the core of most modern businesses’ operations, but it shouldn’t be the starting point of a transformation journey. The goal of digital transformation is to fundamentally enhance business performance and create value: increasing the number of things a business can do (entering new markets, identifying new communities, personalising offers, building new services or products) or improving the way it operates (using insight to support decision making, abstracting away non-core concerns, automating manual work, creating stronger communications loops, predicting demand).
Through this lens, technology is simply an enabler. And companies who get this wrong and task up technology teams when starting a transformation inevitably end up on a simple modernisation journey whose outcomes are typically little more than newer technical implementations. The challenge of justifying this kind of modernisation is that that it is expensive and often the value to the business is tangential – ‘do the same thing a bit more efficiently’.
So if the goal is business value and the enabler is technology, the missing link is capability. A capability is something a business is able to do that can be described discretely, and as a rule of thumb is something you can usually preface with ‘we can..’. One example (there are thousands) is ‘use sensor data to make real-time decisions on the factory floor’.
Building business capabilities can be tough and often spans an intimidating breadth of technical realms: from software engineering to data engineering, data science, business intelligence, cloud platforms, security and DevOps. So how do organisations go about starting to break down and understand this journey, especially when they may be working with a fairly digitally immature base?
If we take our example capability (‘use sensor data to make real time decisions on our factory floor’) we can map out the software and data journey a business must travel through to get there. Each step increases maturity in an area that will additionally pave the way for a range of new capabilities and opportunities.
1. Nail the question.
Before we start implementing a solution, we should first ensure we are answering the right question. When we start with a desired capability such as “Use sensor data to make real time decisions on our factory floor” we should ensure we understand the desired outcome and the implications of this change. Which are the key decisions that are to be made? How does this add value? Outside of the capability to be created what else would need to change to enable the desired outcome (e.g. people, process, tech.)
2. Collect relevant data.
Once we understand the question, we can broadly determine the data that will need to (repeatedly) collect. This is largely a software engineering challenge, interacting with hardware, APIs and building new data input sources. At this point the challenge is largely functional. This is the first step in digital maturity in this space; a factory floor without data cannot be a foundation for digital capability.
3. Store and dimension data.
Data is unwieldy and of little value in its raw form. Understanding the business and the data, and dimensioning and storing it in an efficient and useable manner is a data engineering challenge. Data warehousing and having robust systems to manage data efficiently is the next step towards maturity in this example.
4. Derive insight.
Large volumes of data are hard to understand and turn into meaningful information. Business Intelligence reporting is often the output from such data, but in this example we are interested in automating decisions, not just getting reports. This is a data science challenge. Models must be built and trained using volumes of data a human could not understand. This represents a fairly digitally mature organisation at this point.
5. Take continuous action.
While building a set predictive models with data science can be undertaken as a one-off exercise, the most digitally mature organisations will find huge value in being able to continuously leverage insight that drives action from their data. This is where the promise of cloud meets software and data, and through integrated platforms the output of continuously running models can drive real time outputs into systems: in this case making complex decisions about actions to be taken on a factory floor.
This real-world capability may manifest in different ways in different organisations: a recycling factory may need to use computer vision to better identify and organise waste, while an assembly factory may need to use hardware telemetry sensors to detect machine wear and adjust accordingly – but the maturity steps along the way to achieve those capabilities remain broadly the same.
Another important consideration is that often capabilities build on top of each other. Many larger organisations struggle to tie together different initiatives because they miss the point that many disparate capabilities are often built off the same foundations – which results in fractured data and incompatible technologies over time.
So consider your business in terms of the capabilities it has right now and the capabilities you need to build to deliver value – these should be your north star for making technology decisions. Because ultimately businesses that are able to avoid being distracted by tech for tech’s sake are necessarily closer to the value agenda – and will get much further much faster.